Applicability of Deep Reinforcement Learning for Efficient Federated Learning in Massive IoT Communications
To build intelligent model learning in conventional architecture, the local data are required to be transmitted toward the cloud server, which causes heavy backhaul congestion, leakage of personalization, and insufficient use of network resources. To address these issues, federated learning (FL) is...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-02-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/13/5/3083 |
| _version_ | 1797615730943852544 |
|---|---|
| author | Prohim Tam Riccardo Corrado Chanthol Eang Seokhoon Kim |
| author_facet | Prohim Tam Riccardo Corrado Chanthol Eang Seokhoon Kim |
| author_sort | Prohim Tam |
| collection | DOAJ |
| description | To build intelligent model learning in conventional architecture, the local data are required to be transmitted toward the cloud server, which causes heavy backhaul congestion, leakage of personalization, and insufficient use of network resources. To address these issues, federated learning (FL) is introduced by offering a systematical framework that converges the distributed modeling process between local participants and the parameter server. However, the challenging issues of insufficient participant scheduling, aggregation policies, model offloading, and resource management still remain within conventional FL architecture. In this survey article, the state-of-the-art solutions for optimizing the orchestration in FL communications are presented, primarily querying the deep reinforcement learning (DRL)-based autonomy approaches. The correlations between the DRL and FL mechanisms are described within the optimized system architectures of selected literature approaches. The observable states, configurable actions, and target rewards are inquired into to illustrate the applicability of DRL-assisted control toward self-organizing FL systems. Various deployment strategies for Internet of Things applications are discussed. Furthermore, this article offers a review of the challenges and future research perspectives for advancing practical performances. Advanced solutions in these aspects will drive the applicability of converged DRL and FL for future autonomous communication-efficient and privacy-aware learning. |
| first_indexed | 2024-03-11T07:30:50Z |
| format | Article |
| id | doaj.art-84b8e1e7e67b4474bf1b0a028950f827 |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2024-03-11T07:30:50Z |
| publishDate | 2023-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-84b8e1e7e67b4474bf1b0a028950f8272023-11-17T07:18:59ZengMDPI AGApplied Sciences2076-34172023-02-01135308310.3390/app13053083Applicability of Deep Reinforcement Learning for Efficient Federated Learning in Massive IoT CommunicationsProhim Tam0Riccardo Corrado1Chanthol Eang2Seokhoon Kim3Department of Software Convergence, Soonchunhyang University, Asan 31538, Republic of KoreaDepartment of Information and Communications Technology, American University of Phnom Penh, Phnom Penh 12106, CambodiaDepartment of Software Convergence, Soonchunhyang University, Asan 31538, Republic of KoreaDepartment of Software Convergence, Soonchunhyang University, Asan 31538, Republic of KoreaTo build intelligent model learning in conventional architecture, the local data are required to be transmitted toward the cloud server, which causes heavy backhaul congestion, leakage of personalization, and insufficient use of network resources. To address these issues, federated learning (FL) is introduced by offering a systematical framework that converges the distributed modeling process between local participants and the parameter server. However, the challenging issues of insufficient participant scheduling, aggregation policies, model offloading, and resource management still remain within conventional FL architecture. In this survey article, the state-of-the-art solutions for optimizing the orchestration in FL communications are presented, primarily querying the deep reinforcement learning (DRL)-based autonomy approaches. The correlations between the DRL and FL mechanisms are described within the optimized system architectures of selected literature approaches. The observable states, configurable actions, and target rewards are inquired into to illustrate the applicability of DRL-assisted control toward self-organizing FL systems. Various deployment strategies for Internet of Things applications are discussed. Furthermore, this article offers a review of the challenges and future research perspectives for advancing practical performances. Advanced solutions in these aspects will drive the applicability of converged DRL and FL for future autonomous communication-efficient and privacy-aware learning.https://www.mdpi.com/2076-3417/13/5/3083communication-efficient learningdeep reinforcement learningfederated learningmassive Internet of Thingspolicy optimizationself-organizing networks |
| spellingShingle | Prohim Tam Riccardo Corrado Chanthol Eang Seokhoon Kim Applicability of Deep Reinforcement Learning for Efficient Federated Learning in Massive IoT Communications Applied Sciences communication-efficient learning deep reinforcement learning federated learning massive Internet of Things policy optimization self-organizing networks |
| title | Applicability of Deep Reinforcement Learning for Efficient Federated Learning in Massive IoT Communications |
| title_full | Applicability of Deep Reinforcement Learning for Efficient Federated Learning in Massive IoT Communications |
| title_fullStr | Applicability of Deep Reinforcement Learning for Efficient Federated Learning in Massive IoT Communications |
| title_full_unstemmed | Applicability of Deep Reinforcement Learning for Efficient Federated Learning in Massive IoT Communications |
| title_short | Applicability of Deep Reinforcement Learning for Efficient Federated Learning in Massive IoT Communications |
| title_sort | applicability of deep reinforcement learning for efficient federated learning in massive iot communications |
| topic | communication-efficient learning deep reinforcement learning federated learning massive Internet of Things policy optimization self-organizing networks |
| url | https://www.mdpi.com/2076-3417/13/5/3083 |
| work_keys_str_mv | AT prohimtam applicabilityofdeepreinforcementlearningforefficientfederatedlearninginmassiveiotcommunications AT riccardocorrado applicabilityofdeepreinforcementlearningforefficientfederatedlearninginmassiveiotcommunications AT chantholeang applicabilityofdeepreinforcementlearningforefficientfederatedlearninginmassiveiotcommunications AT seokhoonkim applicabilityofdeepreinforcementlearningforefficientfederatedlearninginmassiveiotcommunications |